pathergy reaction

19 downloads 116 Views 71KB Size Report
myeloid leukaemia (CML) on interferon alpha (IFN-a) therapy and the significance of this observation .... with hepatitis C infection, hairy cell leukaemia, essen-.
British Journal of Rheumatology 1998;37:1148–1151

SKIN HYPERREACTIVITY OF BEHC ¸ ET’S PATIENTS (PATHERGY REACTION ) IS ALSO POSITIVE IN INTERFERON ALPHA-TREATED CHRONIC MYELOID LEUKAEMIA PATIENTS, INDICATING SIMILARLY ALTERED NEUTROPHIL FUNCTIONS IN BOTH DISORDERS ˇ AN, Z. DEMIRC ˇ AN, H. DIRESKENELI,† T. ERGUN,* ¨ . ALPDOG T. BUDAK-ALPDOG ¸ AY,* O ˘ LU ¨ ¨ ¨ ¨ SKENT‡ and T. AKOG A. OZTURK,‡ A. GUNAY,‡ S¸. YAVUZ,† N. U TUBI˙TAK, Behc¸et’s Disease Research Unit, Department of Haematology, *Department of Dermatology, †Department of Rheumatology, Marmara Medical School, I˙stanbul and ‡Department of Haematology-Oncology, Gu¨lhane Military Academy, I˙stanbul, Turkey SUMMARY Typical manifestations of Behc¸et’s disease (BD) and a positive pathergy reaction were observed in a few patients with chronic myeloid leukaemia (CML) on interferon alpha (IFN-a) therapy and the significance of this observation was assessed in a prospective study. The skin pathergy test was applied to 15 patients with CML prior to IFN-a therapy, 29 patients with CML following IFN-a therapy and 30 patients with BD. Twenty-five patients with inflammatory arthropathies (IA), 20 patients with recurrent oral ulcers (ROU ), 23 patients treated with IFN-a for various disorders and 20 normal individuals were also studied as control groups. The pathergy reaction was positive in nearly a quarter of IFN-a-treated CML cases (24%) as well as onehalf of the patients with BD (50%). All CML patients prior to IFN-a treatment and all patients using IFN-a for other diseases were negative for the pathergy reaction. These results may indicate a similarly altered neutrophil function in both BD and IFN-a-treated CML patients. K : Behc¸et’s disease, Chronic myeloid leukaemia, Neutrophil function, Pathergy reaction.

B’ disease (BD) is a systemic disorder presenting with mucocutaneous, ocular, intestinal, vascular and neurological involvement. Mucocutaneous lesions include recurrent oral aphthae, genital ulceration, various skin lesions and a positive pathergy reaction. The pathergy reaction is hyperreactivity of the skin to needle prick and it has been defined as one of the criteria for diagnosis of BD by an International Study Group [1]. Although the prevalence of a positive reaction varies between different series, a positive test has been suggested to be highly specific for BD [2]. The underlying mechanism of skin hyperreactivity in BD is unknown. Histopathological examination of the pathergy reaction reveals a perivascular mononuclear and polymorphonuclear leucocyte (PMNL) infiltration with epidermal thickening, and subcorneal pustule formation [3]. The functions of the PMNLs have been studied in BD, in order to explain the PMNL accumulation in the skin. Chemotactic activity, phagocytosis and random migration of PMNL were found to be increased [4]. Neutrophils obtained from patients with BD were found to damage the normal endothelial cells in culture [5]. Recently, PMNL adhesion to normal endothelium has been shown to be significantly high in BD [6 ]. These findings indicate a significantly high PMNL activity which may be an underlying cause of the skin hyperreactivity to needle prick in this disorder. Chronic myeloid leukaemia (CML) is a malignant

myeloproliferative disorder with an increased number of PMNLs and their precursors in the peripheral blood and in the bone marrow. Circulating neutrophils derived from a CML clone were suggested to fail to reach full cytoplasmic maturation, and they were shown to have reduced complement receptor 1 (CR1) and FcRIII (CD16) expression, which mediate a variety of neutrophil functions [7]. Granulocytes from CML patients have an altered response to different chemoattractants in comparison with normal PMNLs [7]. They also have a decreased NADPH activity, H O -myeloperoxidase-dependent iodination, and 2 2 receptor-mediated endocytosis [7, 8]. Therefore, although there is a significant increase in the number of PMNLs in CML, they have a number of functional abnormalities. Two patients with CML receiving interferon alpha (IFN-a) presented characteristic manifestations of BD: oral and genital aphthae, folliculitis, arthritis and a positive pathergy reaction [9]. These symptoms subsided on withdrawal of the IFN-a therapy. A similar case has been reported from Japan [10] and a further patient was diagnosed at Istanbul University, Istanbul Medical School, who developed BD in the setting of CML whilst on treatment with IFN-a (S. Yavuz, personal communication). In a preliminary survey, the pathergy test was applied to a number of patients with CML prior to and following IFN-a therapy, and two of these patients also developed a positive pathergy reaction following IFN-a treatment. They had no other clinical manifestations of BD [9]. Subsequently, a prospective study was designed in order to reassess the specificity of the pathergy reaction in BD, and to evaluate the effect

Submitted 19 March 1998; revised version accepted 10 June 1998. ¨ niversitesi Tıp Correspondence to: T. Akogˇlu, Marmara U Faku¨ltesi Hastanesi, Hematoloji-I˙mmunoloji Bilim Dali, Tophaneliogˇlu caddesi, 81190 Altunizade, I˙stanbul, Turkey.

© 1998 British Society for Rheumatology 1148

ˇ AN ET AL.: PATHERGY IN IFN-a-TREATED CML BUDAK-ALPDOG

of IFN-a treatment on pathergy positivity in CML patients. PATIENTS AND METHODS Patients and controls The pathergy reaction was assessed in different groups of patients. The first group was selected from the rheumatology out-patient clinics by expert rheumatologists, and consisted of 30 patients with BD fulfilling the International Study Group criteria [1]; 21 patients with classical rheumatoid arthritis (RA) according to ACR criteria, four patients with seronegative spondyloarthropathy and 20 patients with recurrent oral ulcerations (ROU ). Patients with ROU had oral ulcers on at least six occasions per year without other clinical symptoms of BD. The second group consisted of 23 patients who were receiving IFN-a treatment for various illnesses for at least 6 months. These were 10 patients with multiple myeloma in plateau phase, two patients with hairy cell leukaemia, four patients with polycythaemia vera and two patients with essential thrombocythaemia. One patient with renal cell carcinoma, three patients with hepatitis C infection and one patient with malignant melanoma were also studied. The dose of IFN-a for patients with multiple myeloma and renal cell carcinoma was 5 million units, three times a week. Patients with hepatitis C infection, hairy cell leukaemia, essential thrombocythaemia and polycythaemia vera required a dose of 3 million units, three times a week. The patient with malignant melanoma was treated with a relatively more intensive dose of 9 million units per day. The third group consisted of a total 35 CML patients. Fifteen of the 35 CML patients were tested for pathergy reaction prior to IFN-a treatment while they were only treated with hydroxyurea and their total peripheral leucocyte count was below 10 000/ml. Nine out of these 15 patients underwent a second test following 6 months of IFN-a therapy. The remaining 20 patients were on IFN-a treatment for at least 6 months when the pathergy test was applied. Therefore, a total of 29 patients were tested while they were on IFN-a treatment. All CML patients were Philadelphia chromosome positive, and they were all in chronic phase. The total peripheral leucocyte count was below 10 000/ml in all CML patients when the pathergy test was applied. Three million to five million units/m2 of IFN-a were administered daily to all CML patients. The dose was adjusted according to the patient’s tolerance. All patients were administered interferon-a 2b (Schering-Plough, USA). Twenty healthy volunteers were also studied as normal controls. All patients included in the study were also investigated for clinical features of BD. Pathergy test Six needle pricks were made simultaneously (three to each arm) with sharp hypodermic needles of 20G size to the hairless, avascular area of the flexor aspects of the forearm by one dermatologist who evaluated

1149

the reaction after 48 h. The formation of papule or pustule in any one of the six needle pricks was regarded as a positive pathergy reaction. Skin punch biopsy samples were obtained from the positive pathergy reactions of four patients with CML, one patient with RA and one patient with ROU. Fisher’s exact test was employed for comparing the prevalence of a positive pathergy test within groups. The diagnostic indices, such as sensitivity and specificity, and their confidence limits, were calculated according to the guidelines defined by Reid et al. [11]. RESULTS None of the healthy controls and the patients, who were on IFN-a treatment for various malignant and non-malignant disorders other than CML, were observed to have a positive skin pathergy reaction ( Table I ). However, one patient with RA and one patient with ROU were found to have a positive pathergy reaction, which was significantly lower than the prevalence of a positive pathergy test in patients with BD (P < 0.0001). None of these patients had any other features of BD. A positive pathergy reaction was observed in 15 out of 30 (50%) patients with BD. All of the 15 patients with CML who were tested prior to IFN-a treatment had a negative skin prick test while they were receiving only hydroxyurea. However, seven CML patients out of 29 (24%), who were tested whilst they were treated with IFN-a for at least 6 months, had a positive pathergy reaction. Interestingly, five of these seven patients also had oral aphthae after IFN-a therapy. However, no other clinical features of BD were observed in these patients. When the nine patients who were tested twice were assessed separately, two of them were observed to have a positive pathergy reaction following 6 months of IFN-a therapy, although they had a negative reaction prior to IFN-a. Therefore, we suggested that the pathergy reaction could be related to the IFN-a treatment in these cases. Both in the BD group and the CML group on IFN-a therapy, the prevalence of a positive pathergy reaction was significantly higher than the prevalence in the other control groups (P < 0.05 for all ). Biopsy samples from the positive pathergy site revealed both a superficial and a deep perivascular infiltration consisting of mononuclear cells and neutrophils with leucocytoclasis and also intraepidermal pustule formation. The endothelial cells were found to be swollen, but no fibrin deposition could be detected. The histopathological findings were similar for the patients with RA, ROU or CML, and they did not differ from the features of pathergy reaction that had been previously reported for the patients with BD [3]. A pathergy test was found to have 50% sensitivity (95% confidence interval of 32–68%) in cases with BD, while it had 98% specificity (95% confidence interval of 95–100%) upon exclusion of the IFN-a-treated CML group. The specificity of the pathergy test appeared to be 92% (95% confidence interval of

1150

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 37 NO. 11 TABLE I The demographic features of the groups and the results

Groups

n

Median age

F/M

Positive pathergy test

Behc¸et’s disease Inflammatory arthropathies CML before IFN-a treatment Recurrent oral ulcers Healthy controls Diseased control group for IFN-a treatment* CML after IFN-a treatment

30 25 15 20 20 23 29

34 44 35 32 32 62 45

15/15 23/2 9/6 13/7 10/10 6/17 14/15

15/30 (50%) 1/25 (4%) 0/15 1/20 (5%) 0/25 0/23 7/29 (24%)

(21–48) (22–69) (27–64) (18–63) (24–62) (34–82) (24–67)

*Multiple myeloma (10 patients), hairy cell leukaemia (two patients), polycythaemia vera (four patients), essential thrombocythaemia (two patients), and one patient with renal carcinoma, one patient with malignant melanoma and three patients with hepatitis C virus infection.

87–97%) when the CML patients on IFN-a treatment were included. HLA typing was performed in 15 out of 35 patients with CML (11 of them had a negative and four of them had a positive pathergy test) in order to search for a probable donor for allogeneic bone marrow transplantation. None of the CML patients with a positive pathergy reaction had HLA-B51 antigen. DISCUSSION The skin pathergy test was reported to be highly specific for BD [2]. However, the sex and race of the patients, clinical severity of the disease, HLA-B51 status and the method of application of the needle prick were reported to be significant factors in causing a positive reaction [2, 12–15]. The prevalence of the positive pathergy test in patients with BD from Turkey, Japan and Israel was reported to be >50%, in comparison with the healthy or diseased control groups including some haematological disorders, where the prevalence did not exceed 3% [1, 2, 12–15]. This study showed that IFN-a-treated CML patients also had a high prevalence of positive pathergy reaction (24%). Specificity of the test for BD was calculated to be 98% in the groups other than IFN-a-treated CML patients, and 92% if the latter group was included. The underlying reason for the skin hyperreactivity in BD, and the effect of IFN-a treatment on the skin pathergy reaction in CML patients, is unknown. It may be due to altered PMNL activity in both disorders. A number of immunological abnormalities, which may be associated with increased PMNL activities, have been reported in cases with BD. Increased serum levels of IFN-c, increased interleukin-2 (IL-2) production in response to non-specific stimulation, increased soluble IL-2 receptor concentrations, and significantly increased production of tumour necrosis factor alpha ( TNF-a), IL-6 and IL-8 by the mononuclear cells were reported in patients with BD [16 ]. Different investigators, including our group, also showed that patients with BD had significantly increased PMNL functions [4–6 ]. Endothelial adhesion of neutrophils, chemotaxis, and random migration and phagocytosis were reported to be higher in BD patients [4–6 ]. A number of abnormalities of PMNLs have been reported in the patients with CML. Some of these abnormalities were shown to be reversible with IFN-a

treatment [7, 8]. In a comparative study, we have recently shown that adhesion of PMNLs to a glass surface was significantly higher in patients with BD and CML compared with the PMNLs obtained from normal controls. However, there was a difference between the two diseases in that a higher activity was observed in patients with BD. Interestingly, in vitro incubation of PMNLs from CML patients with IFN-a caused an increment of glass adhesion to the levels that were observed with the PMNLs from BD patients ( Karti et al. unpublished data). IFN-a has multiple effects on the immune system, including activation of monocyte/macrophages. Following this activation, important inflammatory cytokines such as IL-1a, IL-1b, TNF-a, interferons a, b, c, and IL-6 are secreted [17]. IFN-a treatment of normal volunteers is associated with increased plasma levels of IL-6, IL-8 and IL-10 [18]. More importantly, increased neutrophil activity was detected in these volunteers following IFN-a treatment, as observed by the increment in the plasma concentrations of elastasea1–anti-trypsin complexes and lactoferrin [18]. In addition to general effects on the mature neutrophils, IFN-a also affects the CML clone. These cells are deficient for some adhesion molecules and present functional abnormalities of the b1-integrins [19]. IFN-a treatment was observed to reverse the abnormal adhesion characteristics of the CML clone [19]. We propose that increased skin reaction to needle prick in both CML and BD may reflect altered PMNL activities. In CML patients, abnormal PMNL functions are a consequence of the malignant clone, and IFN-a treatment possibly modulates them to act in a similar fashion to the PMNLs of Behc¸et’s patients. Increased PMNL activity in BD may be due to the significantly increased inflammatory mediators in the plasma of the patients [16 ], or it may be a consequence of a constitutive abnormality in the neutrophils in this disorder. Although a genetic locus responsible for BD could not be identified, linkage with HLA-B51 was suggested by different investigators [16 ]. Indeed, it has been reported that HLA-B51-positive normal individuals, as well as HLA-B51 transgenic mice, had increased neutrophil functions [20, 21]. Neutrophils obtained from the Behc¸et’s patients, both in active state and remission, show similar features with respect to their activity. Finally, IFN-a treatment results in

ˇ AN ET AL.: PATHERGY IN IFN-a-TREATED CML BUDAK-ALPDOG

pathergy positivity in a subgroup of CML patients, and this provides further evidence that neutrophil activity in BD may be caused by an inherent abnormality.

10. 11.

R 1. The International Study Group for Behc¸et’s Disease. Criteria for diagnosis of Behc¸et’s disease. Lancet 1990;335:1078–80. ¨ cal L, Inanc¸ M, Gu¨l A. 2. Dilsen N, Konic¸e M, Aral O, O Comparative study of the skin pathergy test with blunt and sharp needles in Behc¸et’s disease: Confirmed specificity but decreased sensitivity with sharp needles. Ann Rheum Dis 1993;52:823–5. 3. Ergun T, Gu¨rbu¨z O, Harvell J, Jorizzo J, White W. The histopathology of pathergy: a chronological study of skin hyperreactivity in Behc¸et’s disease. Int J Dermatol; in press. 4. Mizushima Y. Chemotaxis and phagocytosis of leukocytes in Behc¸et’s disease. An overview. In: Lehner T, Barnes CG, eds. Recent advances in Behc¸et’s disease. Royal Society of Medicine Services, London, 1986:85–7. 5. Niwa Y, Miyake S, Sakane T, Shingu T, Yokoyama M. Auto-oxidative damage in Behc¸et’s disease—endothelial cell damage following the elevated oxygen radicals generated by stimulated neutrophils. Clin Exp Immunol 1982;49:247–55. 6. S¸ahin S¸, Akogˇlu T, Direskeneli H, S¸en LS, Lawrence R. Neutrophil adhesion to endothelial cells and factors affecting adhesion in patients with Behc¸et’s disease. Ann Rheum Dis 1996;55:128–33. 7. Radhika V, Thennarasu S, Naik NR, Kumar A, Advani SH, Bhisey AN. Granulocytes from chronic myeloid leukemia CML patients show differential response to different chemoattractants. Am J Hematol 1996; 52:155–64. 8. Anklesaria PN, Advani SH, Bhisey AN. Studies on granulocyte functions in patients with chronic myeloid leukemia. Tumori 1985;71:317–23. ¨ et al. 9. Budak-Alpdogˇan T, Demirc¸ay Z, Alpdogˇan O Behc¸et’s disease in patients with chronic myelogenous leukemia: possible role of interferon-alpha treatment in

12.

13. 14.

15.

16. 17.

18.

19. 20. 21.

1151

the occurrence of Behc¸et’s symptoms. Ann Hematol 1997;74:45–8. Segawa F, Shimizu Y, Saito E, Kinoshita M. Behc¸et’s disease induced by interferon therapy for chronic myelogenous leukemia. J Rheumatol 1995;22:1183–4. Reid MC, Lachs MS, Feinstein AR. Use of methodological standards in diagnostic test research. Getting better but still not good. J Am Med Assoc 1995;274:645–51. Yazıcı H, Chamberlain MA, Tu¨zu¨n Y, Yurdakul S, Mu¨ftu¨ogˇlu A. A comparative study of the pathergy reaction among Turkish and British patients with Behc¸et’s disease. Ann Rheum Dis 1984;43:74–5. Friedman-Birnbaum R, Bergman R, Aizen E. Sensitivity and specificity of pathergy test results in Israeli patients with Behc¸et’s disease. Cutis 1990;45:261–4. Yazıcı H, Tu¨zu¨n Y, Pazarlı H, Yalc¸ın B, Yurdakul S, Mu¨ftu¨ogˇlu A. The combined use of HLA-B5 and the pathergy test as diagnostic markers of Behc¸et’s disease in Turkey. J Rheumatol 1980;7:206–10. Fresko I, Yazıcı H, Bayramic¸li M, Yurdakul S, Mat C. Effect of surgical cleaning of the skin on the pathergy phenomenon in Behc¸et’s syndrome. Ann Rheum Dis 1993;52:619–20. Emmi L, Salvati G, Brugnolo F, Marchione T. Immunopathological aspects of Behc¸et’s disease. Clin Exp Rheumatol 1995;13:687–91. Sreevalsan T. Biologic therapy with interferon-a and b: preclinical studies. In: De Vita VT, Hellman S, Rosenberg SA, eds. Biologic therapy of cancer. Philadelphia: J.B. Lippincott, 1995:347–65. Corssmit EPM, Heijligenberg R, Hack CE, Endert E, Sauerwein HP, Romijin JA. Effects of interferon-alpha (IFN-a) administration on leukocytes in healthy humans. Clin Exp Immunol 1997;107:359–63. Cervantes F, Rozman C. Benign hematopoietic progenitors in chronic myeloid leukemia: current status and future prospects. Ann Hematol 1994;69:99–105. Sensi A, Gavioli R, Spisani S et al. HLA B51 antigen associated with neutrophil hyper-reactivity. Dis Markers 1991;9:327–31. Takeno M, Kariyone A, Yamashita N et al. Excessive function of peripheral blood neutrophils from patients with Behc¸et’s disease and from HLA-B51 transgenic mice. Arthritis Rheum 1995;38:426–33.